Wi-Fi Probe Request Fingerprinting for Unambiguous Device Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods fail to unambiguously identify Wi-Fi devices not connected to a Wi-Fi network, especially when they employ MAC address randomization, as they rely on fixed MAC addresses and do not account for devices with variable fields in probe request frames.
Innovation Solution
A method that extracts distinctive fields from probe request frames, projects them into a feature space, and uses machine-learning techniques to create unique footprints or signatures for unconnected Wi-Fi devices, allowing identification even with random MAC addresses, and stores this information for further analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If MAC address randomization is used to prevent tracking, then user privacy is protected, but device identification becomes ambiguous
Solution Approach 1:
The patent segments the probe request frame into multiple distinctive fields (capability information, supported rates, HT capabilities, VHT capabilities, vendor-specific information) rather than relying on a single MAC address. Each field is extracted and analyzed separately to build a composite device fingerprint that remains consistent even when MAC addresses are randomized.
Solution Approach 2:
The patent creates a unique 'color' or signature for each device by combining multiple field characteristics from probe request frames. This composite identifier acts like a unique visual signature that remains consistent across different MAC addresses, enabling reliable identification while preserving privacy.
2Adaptability or versatility
If variable fields in probe request frames are used for identification, then more devices can be identified, but complexity of analysis increases
Solution Approach 1:
The patent extracts specific distinctive fields from the probe request frame structure, separating the identification-relevant information from the variable MAC address fields. By taking out and analyzing only the stable capability information fields (supported rates, HT capabilities, VHT capabilities, vendor-specific), the system achieves versatile device identification without analyzing the entire frame, thus managing complexity.
Solution Approach 2:
The patent transitions from single-dimensional MAC address identification to multi-dimensional identification by analyzing multiple fields simultaneously (capability information, supported rates, HT capabilities, VHT capabilities, vendor-specific information). This dimensional expansion enables identification of diverse devices while the structured approach to analyzing each dimension separately keeps the overall system manageable.
3Measurement precision
If multiple fields are extracted and analyzed, then identification accuracy improves, but processing time increases
Solution Approach 1:
The patent performs preliminary extraction and categorization of distinctive fields from probe request frames, organizing the information into structured components before detailed analysis. By pre-identifying and separating the relevant fields (capability information, supported rates, HT capabilities, VHT capabilities, vendor-specific), the system reduces the computational burden of subsequent analysis, achieving high identification accuracy without excessive processing time.
Data Source
AI summary
A device and method for the unambiguous identification of Wi-Fi devices is disclosed, by receiving probe request frames sent by Wi-Fi devices, extracting a set of features from a plurality of fields of each probe request frame, assigning a footprint to each probe request frame based on the extracted set of features, for each footprint, performing a cluster analysis on a time series of the sequence numbers included in the header of the probe request frames associated with the corresponding footprint, obtaining at least one cluster per footprint, identifying a Wi-Fi device for each different cluster, which identifies Wi-Fi devices that are not connected to a Wi-Fi network and employ MAC address randomization.